Connectivity Providers Database

From: JJTNPRV@grouch.jpl.nasa.gov To: Multiple recipients of list PLEARN-L <PLEARN-L@ubvm.cc.buffalo.edu> Date: Sun, 23 Aug 1992 21:38:14 -0700 >Let me add a minor point to this. The transition to a workstation base may mean >transition to "intel based workstations" (486/586) and NT operating system. >Engineering workstations (Sun, HP, IBM, SGI) certainly have their place in the >market today, but tomorrow may look a bit different. >In terms of price/performance these "intel based workstations" are getting clos e >to the engineering workstations. And a brief look at the hardware base makes on e >wonder about a future of Un*x (vs NT): last year "the world" bought 27 million >PCs and 500,000 workstations. >Thus, I would not read Mr. Goldstein's words as the indication that the future >of networking in EEurope should be equated with a traditionally defined >engineering workstations. Having returned from Poland This last Thursday, I have a few fresh words to say about Polish networking. Poland, and this is in general, and not just networking, is a land of contrast now. In Warsaw you have 2% unemployment like in Japan, and in DolnoSlask you have 20% (an order of magnitude difference). This order of magnitude difference also applies to universities. Mr. Goldstein visited Warsaw and Krakow. These are the prime sites. I visited Maciek Uhlig in Katowice, University of Slask. This is the backwaters of networking. The biggest problem with the internet in US is to convince the elder professors that they won't be humiliated or loose their jobs if they connect to the network and find out that the same thing that they are doing has already been done, or is being done better elsewhere. Also, the prized trips to the West can be in jeopardy if the work can be done remotely over the internet. The internet at US consists of about 2-3 PC's and a thin wire ethernet. But Maciek Uhlig is great, and he has alot of enthusiasm about system administration and the internetwork. The bottom line I told him is that you don't need a fancy shmansy AGS or AT500 if you're good. We had what, 1/5 of a 9600bps PTT channel linking us to the outside world? NO PROBLEMO I said. Within oh, 5-10 minutes, we were doing X Windows/Motif applications over the internet from my client machines at JPL in Pasadena on his $2000 ALR 486 system running SCO Unix. He told me about the big and great universities like Krakow, which he'd like to work for, which use FDDI. I asked him how many PC's he needs to connect? He said his goal is 300. I told him to install a thick wire ethernet cable with maybe at most a CISCO IGS or PC/XT running KA9Q or two connected to a multiport repeater with 8 thin wire segments, one for each floor of the buildings he needs to connect. I suggested the inexpensive, good, solid, proven track record approach, because the bottom line is that he doesn't need an FDDI configuration. My biggest surprise? The 9600bps PTT line was quite reliable. I was expecting a dog connection. A constant source of headaches. Not so. His major greivance is cabling expertise. I told him about half a year ago to watch out for cabling. You have to know how to make the taps. You have got to know what you're doing with the cables. Otherwise there may be deviling problems later. It seems that the companies don't exist in the region that know what they're doing with cables. It was sort of my explanation to him, which he took quite awkwardly, that as an educational center it was their responsibility to build this type of expertise in the community. I suggested they contract with a local firm, and part of the contract would be to send people to cabling workshops in the West. Then this company should come back and do a job on their network. This I explained is what NASA does with contractors. Sometimes the company is less advanced than you are, and then you have to provide it with the ways and means to serve you. Until they become more advanced than you are in the field. What Krakow did, I consider a cop-out. They hired CISCO and ALLIED TELESIS to set the FDDI network up for them. And they're here bragging about their network to SG. This does not produce local skills. But maybe it worked, so who's to say it's wrong. Anyway, I travelled all the way to Katowice from Miedzylesie and back by train to see this big network (2-3 PC's) but it was worth the experience. In the future, M. Uhlig and I will be communicating by means of xfig graphs. This PC business is better because it is alot more cost effective for the little guy universities. JJT